LongShOT Leaderboard

48 models

Rankings

Model Leaderboard

Performance rankings across 16 task categories on LongShOTBench. All scores in %.

Last updated: March 30, 2026

Qwen3-Omni-30B-A3B-Thinking

T130B (3B active)

61.52%

49.3

71.6

info

62.8

62.5

Qwen3-VL-32B-Thinking

T332B

45.66%

36.7

52.6

info

43.7

49.6

Qwen3-VL-30B-A3B-Thinking

T330B (3B active)

42.40%

34.2

48.8

info

39.9

46.6

CP Core Perception TasksRE Reasoning TasksInfo Information TasksMM Multimodal Tasks

				Score	Category				Modality			Duration
#	Model	Params	Input	Overall	CP	RE	Info	MM	Visual	Audio	Speech	Short	Med	Long
	Qwen3-Omni-30B-A3B-Thinking Native Video + Audio	30B (3B active)	V+A	61.52	49.26	71.59	62.77	62.45	59.99	53.69	61.77	58.64	60.75	58.17
	Qwen3-VL-32B-Thinking Mid-Size Video VLMs (10B–50B)	32B	Video	45.66	36.73	52.62	43.70	49.61	43.63	42.42	44.45	42.48	42.85	45.37
	Qwen3-VL-30B-A3B-Thinking Mid-Size Video VLMs (10B–50B)	30B (3B active)	Video	42.40	34.21	48.82	39.92	46.64	40.21	38.74	41.06	38.58	39.33	42.27
4	Qwen3-VL-8B-Thinking Small Video/Vision Models (5B–10B)	8B	Video	41.68	32.90	47.74	38.70	47.39	38.70	37.37	39.45	41.47	37.88	40.46
5	Intern-S1-mini Mid-Size Video VLMs (10B–50B)	14B	Video	39.85	31.74	45.52	36.89	45.26	37.24	35.37	37.97	36.76	36.30	39.45
6	Gemma-3-27B-IT Mid-Size Video VLMs (10B–50B)	27B	Frames	39.65	30.78	48.26	36.55	43.02	37.15	35.08	37.96	39.94	36.17	39.09
7	Gemma-3-12B-IT Mid-Size Video VLMs (10B–50B)	12B	Frames	38.19	29.90	45.36	34.12	43.37	35.26	33.21	36.09	33.81	34.20	37.63
8	Qwen3-VL-4B-Thinking Compact Models (≤5B)	4B	Video	36.26	28.57	41.35	33.08	42.03	33.64	32.23	34.21	35.22	32.93	35.12
9	Qwen3-VL-32B-Instruct Mid-Size Video VLMs (10B–50B)	32B	Video	35.99	29.54	40.29	32.48	41.67	33.52	31.97	34.00	38.05	33.22	33.88
10	MiMo-VL-7B-RL-2508 Small Video/Vision Models (5B–10B)	7B	Video	35.12	26.35	41.58	31.13	41.42	32.10	30.10	32.77	31.33	31.33	33.79
11	Gemma-3-4B-IT Compact Models (≤5B)	4B	Frames	34.38	25.87	40.11	30.55	40.97	31.16	30.10	31.91	34.34	30.14	33.34
12	Qwen3-Omni-30B-A3B-Instruct Native Video + Audio	30B (3B active)	V+A	33.76	25.64	42.59	30.24	36.58	30.72	27.08	31.65	22.60	30.65	31.38
13	Qwen2.5-VL-32B-Instruct Mid-Size Video VLMs (10B–50B)	32B	Video	31.36	26.16	35.23	28.55	35.51	29.50	27.04	29.96	31.39	28.57	31.53
14	GLM-4.6V-Flash Mid-Size Video VLMs (10B–50B)	9B	Video	30.46	22.74	34.75	26.67	37.69	27.20	25.33	27.77	29.44	26.08	29.63
15	Qwen3-VL-8B-Instruct Small Video/Vision Models (5B–10B)	8B	Video	30.01	24.38	33.17	25.02	37.46	26.64	25.30	27.05	32.80	26.01	27.86
16	Ovis2.6-30B-A3B Mid-Size Video VLMs (10B–50B)	30B (3B active)	Video	29.99	22.00	33.19	26.92	37.85	27.37	25.37	27.91	32.80	26.86	28.18
17	Qwen3.5-35B-A3B Mid-Size Video VLMs (10B–50B)	35B (3B active)	Video	26.86	17.65	32.59	22.13	35.07	22.60	20.24	23.23	23.78	22.00	23.91
18	Qwen3.5-27B Mid-Size Video VLMs (10B–50B)	27B	Video	26.64	18.19	33.27	23.04	32.04	23.70	20.60	24.32	23.24	22.88	25.58
19	Qwen2.5-VL-72B-Instruct Large Video VLMs (50B+)	72B	Video	26.34	20.54	30.09	22.80	31.91	24.15	21.39	24.60	24.66	23.24	26.17
20	Qwen3-VL-30B-A3B-Instruct Mid-Size Video VLMs (10B–50B)	30B (3B active)	Video	25.82	21.53	27.91	22.46	31.37	23.87	23.15	24.26	29.50	23.32	24.81
21	InternVL3-78B Large Video VLMs (50B+)	78B	Video	23.58	18.47	25.12	21.16	29.58	20.89	18.71	21.21	25.07	20.31	21.80
22	InternVL3-38B Mid-Size Video VLMs (10B–50B)	38B	Video	23.48	19.29	26.79	20.07	27.75	21.14	18.92	21.49	26.08	20.65	22.13
23	Qwen3-VL-4B-Instruct Compact Models (≤5B)	4B	Video	23.34	18.33	25.42	19.37	30.25	20.67	19.13	21.00	29.03	20.10	21.71
24	Qwen3.5-9B Small Video/Vision Models (5B–10B)	9B	Video	23.30	14.63	29.24	19.88	29.47	19.05	17.28	19.55	21.18	18.53	20.15
25	Ovis2.5-9B Small Video/Vision Models (5B–10B)	9B	Video	23.19	17.50	25.84	19.16	30.26	20.31	18.76	20.60	25.01	19.69	21.50
26	Molmo2-8B Small Video/Vision Models (5B–10B)	8B	Video	23.10	17.52	24.72	18.78	31.38	20.12	18.72	20.32	28.61	19.77	20.43
27	InternVL3-14B Mid-Size Video VLMs (10B–50B)	14B	Video	21.21	17.35	23.75	18.56	25.19	19.69	17.56	19.98	20.06	19.07	21.02
28	InternVL3.5-8B Small Video/Vision Models (5B–10B)	8B	Video	21.10	15.32	21.49	17.57	30.02	17.97	16.62	18.17	21.00	17.48	18.90
29	LongVT-RL Small Video/Vision Models (5B–10B)	7B	Video	20.91	16.44	21.41	18.08	27.73	18.81	18.22	19.03	21.83	18.31	19.73
30	InternVL3.5-38B Mid-Size Video VLMs (10B–50B)	38B	Video	20.73	15.80	22.06	16.41	28.65	18.42	17.22	18.66	21.42	17.76	19.76
31	InternVL2.5-78B Large Video VLMs (50B+)	78B	Video	20.38	16.40	21.97	16.30	26.87	17.45	16.17	17.74	20.06	16.92	18.52
32	Ovis2-16B Mid-Size Video VLMs (10B–50B)	16B	Frames	19.84	15.00	22.43	15.09	26.86	17.44	16.22	17.75	19.94	17.07	18.17
33	InternVL3-8B Small Video/Vision Models (5B–10B)	8B	Video	18.94	15.36	18.92	15.97	25.52	16.88	15.55	17.00	20.00	16.45	17.68
34	InternVL2.5-38B Mid-Size Video VLMs (10B–50B)	38B	Video	18.86	15.96	20.31	14.13	25.02	16.87	15.97	17.17	17.29	16.25	18.23
35	Qwen2.5-VL-7B-Instruct Small Video/Vision Models (5B–10B)	7B	Video	18.39	15.26	20.44	17.00	20.84	17.75	16.53	18.04	20.47	17.32	18.60
36	Kimi-VL-A3B-Thinking Mid-Size Video VLMs (10B–50B)	8B (3B active)	Frames	17.89	11.52	20.94	14.48	24.61	14.85	13.37	15.20	15.28	14.66	15.11
37	Qwen3.5-4B Compact Models (≤5B)	4B	Video	17.78	11.49	23.02	15.40	21.20	15.49	14.39	15.85	16.64	15.50	15.23
38	InternVL3.5-4B-Instruct Compact Models (≤5B)	4B	Video	17.11	13.84	18.02	13.39	23.20	15.33	13.75	15.57	13.39	14.90	16.43
39	Keye-VL-1.5-8B Small Video/Vision Models (5B–10B)	8B	Video	17.09	12.81	19.26	14.33	21.94	15.12	14.56	15.37	18.29	14.55	16.24
40	MiniCPM-o 4.5 Native Video + Audio	8B	Video	17.05	15.08	20.05	16.92	16.13	0.00	0.00	0.00	0.00	0.00	0.00
41	MiniCPM-V 4.5 Small Video/Vision Models (5B–10B)	8B	Video	16.29	12.54	17.84	12.73	22.06	14.01	13.87	14.29	14.93	13.53	15.13
42	Qwen2.5-Omni-7B Native Video + Audio	7B	V+A	15.89	12.49	16.92	14.36	19.81	14.75	12.97	15.08	17.17	14.73	14.81
43	Ovis2-8B Small Video/Vision Models (5B–10B)	8B	Frames	15.16	11.61	16.26	12.04	20.74	12.92	12.51	13.10	12.80	12.30	14.36
44	MiniCPM-o 2.6 Native Video + Audio	8B	Video	14.41	11.27	14.36	11.78	20.24	0.00	0.00	0.00	0.00	0.00	0.00
45	Ovis2-4B Compact Models (≤5B)	4B	Frames	13.07	9.82	13.47	10.18	18.83	11.04	10.82	11.09	10.50	10.34	12.63
46	Kimi-VL-A3B-Instruct Mid-Size Video VLMs (10B–50B)	8B (3B active)	Frames	11.35	8.12	11.23	8.12	17.92	9.37	9.15	9.56	8.85	9.23	9.65
47	LLaVA-OneVision-7B Small Video/Vision Models (5B–10B)	7B	Video	9.14	7.20	8.86	7.58	12.92	7.91	7.89	8.06	9.91	7.85	7.95
48	Qwen2-Audio-7B-Instruct Audio-Only LLMs	7B	Audio	8.07	6.04	7.61	6.16	12.47	7.00	6.88	7.14	6.19	6.64	7.92